A Model Thrombus System for Assessing Thrombolysis

Abstract number: P0787

Mutch NJ, Moore NR, Booth NA

Thrombolysis in vitro is often measured on plasma clots, which are a poor model for thrombi formed in vivo. We have developed a model thrombus system, in which thrombi are formed under flow, such that they have distinct heads, rich in platelets and cells, and fibrin-rich tails. FITC-fibrinogen is incorporated in the thrombi and fluorescence release measures the extent of lysis. We have used the model extensively to examine spontaneous lysis in the thrombus heads and the effects of different physiological inhibitors on lysis by tPA, uPA or scuPA. Lysis in most of these applications is achieved in a buffer milieu. Here we have adapted the system by bathing model thrombi in 80% autologous plasma. The plasma concentration was chosen to be as high as possible but with the capacity to add thrombolytic agents and other drugs or proteins to be tested for potential interactions. It was practicable to prepare about 24 model thrombi, each made from 0.9 L of citrated blood, from each donor, providing sufficient replicates to assess several experimental variations. Lysis by added tPA, 0, 0.2, 1 or 5 g/ml (0–70 M) was followed for 5. Data were analysed by fitting them to appropriate equations, which allowed quantitative interpretation. Lysis was initially most efficient with the highest concentration of tPA but this reached a plateau, and could be overtaken by lower concentrations of tPA. This is consistent with the well-known phenomenon of ‘plasminogen steal’. Lysis was therefore assessed with added plasminogen, both incorporated in the model thrombus and in the bathing fluid. This only improved lysis to a small extent and plasmin-a₂-antiplasmin complex production in the bathing fluid was also elevated. This model system has advantages of being relevant, simple, non-radioactive and quantifiable, characteristics that make it ideal for studying potential interactions with thrombolytic agents.